Main Article Content
Various efforts have been made to reduce drilling costs in the oil and gas upstream industry. One of it is by maximizing drilling Rate Of Penetration (ROP), the speed at which a drill bit breaks the formation underneath it to deepen the borehole. High ROP resulted in shorter drilling times can reduce drilling costs. This is the ideal condition that is expected in every drilling process. However, many factors such as environmental factors (rock formations, wellbore size, drilling mud), drilling parameters (weight on bits, rotational speed, flow rate, hydraulics, etc.) and the characteristics of the bits determine the ROP. Among all, drilling parameters is the only one that can be customized to generate the highest ROP during the drilling process. Choosing drilling parameters to generate the highest ROP in the various environmental condition is not a trivial thing. Moreover, the correlation among these parameters is not linear, and some other factors also affect ROP. Some empirical ROP models that can be used requires parameters that are not always available in the operation field. This study proposes an Artificial Neural Network (ANN) to predict ROP. Using formation type and drilling parameters data as the input, the model produces a great degree of accuracy (R-square at least 0.8). It shows that ANN can become a better alternative to find the optimum drilling parameter to achieve the highest ROP.